Welcome to od|forum

Register now to gain access to all of our features. Once registered and logged in, you will be able to contribute to this site by submitting your own content or replying to existing content. You'll be able to customize your profile, receive reputation points as a reward for submitting content, while also communicating with other members via your own private inbox, plus much more! This message will be removed once you have signed in.


  • Content count

  • Joined

  • Last visited

Community Reputation

3 Neutral

About Pancho

  • Rank

Personal Information

  • Name Tom
  • Location Germany

Recent Profile Visitors

1,554 profile views
  1. P.P.S.: There are other examples out there, using several high res trails which are then combined for rendering in order to get the detail. One sim might not be enough.
  2. First of all, your resolution is obviously too low, much to low. And I guess you miss a point. This is an explosion. Debris is flying around, but the main pressure is coming from the center of the explosion. The trails are only a part of it. If you watch closely you'll realize that the video you posted consists of two elements, trails (kind of low res) and a center cloud which expands rapidly. P.S.: Since the image is a still, don't forget that it is heavily photoshoped.
  3. Just received help on the Redshift forum. Thanks again! "For this type of rendering you need to disable “Non-Blocking Current Frame Rendering” on the RS ROP."
  4. Since I need the sim for a still, the actual behaviour isn't so important than the end position. So I managed to "control" the jello with a drag node set to 0.85 for v and torque. But this slows down the overall animation. Which attribute "inside" the grain node can I alter in a similar fashion, but without changing the amount of other forces. Is there a value which stores the kinectic energy and can be reduced with a drag node? Somehow a force remains inside the grains. Otherwise they couldn't jump out of the glass after they've been stable for a few frames. THere's something like a spring force which makes everything explode.
  5. I tried to figure it out by myself, but don't get it. I've read that you either need to merge all ROPs and render the merge node or connect all ROPs and render the last one. Neither works. It always renders the first take and then stops. I render print res with the Indie version, so "render current frame" is the only option in terms of choosing which frame I want to render. I use Redshift. Did anybody encounter similar problems? Running to the PC every few minutes and check whether one can manually start rendering the next take is not a real life option (though the only one which currently "works"). Cheers Tom
  6. I went from 20 to 100 with the Constraint Iterations which is definitely enhancing the shape conservation, but I haven't figured out how to take the energy away from the grain. I let several layers of fruit fall into my glass. At one point they jump all out of it. Although they were all down at the bottom of the glass. But as soon as the last layers dropped onto the pile somehow all the inital energy gets added up and they jump together out of the glass. NOT - SO - GOOD.
  7. I spent hours today simulating some fruit falling into a glass with FEM. I took forever to calculate and I still ended with horrible penetrations of the glass. I switched over to granular solids and it is soooo much faster, BUT the wiggling and jittering is an issue. A big one. it's like a rodeo which doesn't stop. How do I take away the energy from the sim effectively without going for a diiferent look? A goo like material should be as bouncy as a rocksolid one, and the other way round. Both should come to an abrupt stop with the first collision. Whatever is needed. So, what's the secret? Cheers Tom P.S.: Even values of 10.000.000 for both stiffness parameters don't prevent a heavy deformation on the first collision. Is it not possible to go a bit more into the rigid direction? The additional parameters are too time consuming. Not really possible to use them.
  8. Thanks for the quick reply! I guess the problem with this approach would be if you look at a flat angle into the object in the clear area, it'll remain clear. It won't become opaque towards the opaque area since the material influence is defined at the outside. This approach works great, as long as you keep the view from the side, since it doesn't reveal that the textures aren't volumetric.
  9. I need to layer a colored/tranparent material with a milky one, e.g. a layer of honey and yoghurt in a glass on top of each other. I guess there are are three ways to do that. a) Having one geometry with two different textures assigned. PRO - The transition can be smooth, since both materials can be mixed CON - If I look through the transparent texture into the area/region of the milky one, it'll remain transparent since this isn't a volumetric approach. The textures will be assigned on the outside, so no change inside of the material. Having one geometry for each material PRO - The texture will always change at the boundary of each material. Looking through the honey into the yoghurt works. CON - It's an abrupt change from one material to the other, no gradient. Unnatural look. c) Having a volumetric approach. BUT how would it work? With VDBs or with particles carrying the information on how much each material is represented at this location? Wouldn't this keep me from rendering with the regular materials (Redshift)? I wonder if this works at all? Anybody an idea??? Cheers Tom
  10. Not sure why the group node got dissected into several sub nodes. Unfortunately my $NVTX > 4 expression isn't working anymore. Seems that I'm forced to use a vex expression. The only one which might do the trick is primvertexcount, but I don't get the context right. I'm using the Group expression node and "primvertexcount > 4" isn't working. Anyone?
  11. This seems to be H16, not hardware specific. H15.5 runs smoothly.
  12. Cool! Thanks for the clarification. So, then waiting for new builds and hope the lags will be gone is the only way to get a usable H16?
  13. Thanks for the hint, but I still don't understand whether I need the regular H16 version or the one with support for Qt4. So far there's none with Qt5 support. Will there be less lag with Qt4 support? If the Qt4 version is the one to go, where do I get Qt5 from? The question I get in this link, give me the impression that I want to download a developer's toolkit. No idea at all what this is all about/correct. https://www.qt.io/download/ P.S.: Running Windows 7 on all workstations which aren't connected to the web.
  14. Thanks, but which version do I need? No explanation on the download site.